Navigating the New Reality in DC: An EIP Primer
Last updated: March 2025
Twice now, the Effective Institutions Project has surveyed our network of experts to answer a simple question: what are the most important institutions in the world? Both times — the first in the lead-up to our 2022 landscape analysis, and this past summer for our forthcoming update of that piece — the top answer has been the US federal government. It’s not just Americans that answer this way, or national security experts, or people concerned about risks from AI. We get the same answer from global development professionals, animal welfare advocates, and experts across the Global South. Overall, in our most recent survey, nearly half of respondents named some combination of the US President and/or Congress among their top five picks.
And for good reason: decisions made across the executive, legislative, and judicial branches not only directly impact the lives of 330 million Americans, they resonate well beyond the country’s borders, shaping challenges from emerging technologies to global health. Arguably, at no time in history will the stakes of US government decisions be higher for the world than during the next four years. A Republican majority in Congress and a conservative Supreme Court could provide the administration with a potentially clear path to reshape the structure of American government, with far-reaching consequences for democracy, public welfare, and global stability.
But just because rapid, large-scale change is a possibility by no means guarantees it will become reality—nor is it a foregone conclusion that all of that change will be for the worse. If there’s one theme we are hearing over and over again from our colleagues on the ground, it’s that the Trump administration is likely to be high-variance. The chaotic nature of Trump’s management style and the loose coalition of interests that make up his inner circle create a dynamic environment where wild ideas can travel long distances and decisions can be made and reversed at a moment’s notice. Moreover, though conservative in label, in practice the Trump administration is deeply distrustful of the status quo and the elite institutions (including within government) that maintain it. We can therefore anticipate a strong theme of creative destruction as the administration proceeds, which alongside very real risks could also create unexpected windows of opportunity to improve government functioning and upend outdated or harmful bureaucratic norms.
In EIP’s view, this means that trying to guide what happens in the Trump administration is a significant opportunity, although the durability of impact—especially from disasters averted—will be hard to count on. Wise leadership during this period will involve acting decisively to safeguard critical government infrastructure and checks and balances while redirecting the winds of change to blow where change was most needed all along.
Possible Futures and What We’re Watching
EIP’s north star is to try to make sure that the world’s most powerful institutions are making decisions that help people live happier and more fulfilling lives, both today and in the future. Our work is rooted in the evidence scientists have gathered for decades about what kinds of life events and circumstances impact self-reported wellbeing and life satisfaction, and analysis about how institutional decisions contribute to those outcomes. As part of that analysis, we try to imagine “plausible best- and worst-case scenarios” for institutional performance, both in general and in relation to key issue areas that seem especially consequential for wellbeing. For the US government over the next four years, we are tracking scenarios across five themes right now: AI governance, peace & security, democracy and institutional accountability, global health (including biosecurity), and state capacity. Later in 2025, we’ll add scenarios for animal welfare, climate change, and global development, and update this post accordingly.
With each scenario, we offer our prediction of how things are going to go on a scale of 0 to 10 based on the current signals we’re seeing, with 1 representing our listed worst-case and 9 representing our best-case scenario. (We reserve 0 for “worse than worst” and 10 for “better than best” respectively, to account for the intrinsic uncertainty of the exercise and the fact that sometimes the universe just throws you curveballs.) Remember that these scores are assessing outcomes relative to expectations — so in other words, a high score does not necessarily mean that we think the administration is having a large positive impact in absolute terms or vice versa.
Artificial intelligence and emerging technologies
(You can read about why we think AI governance is important and our takes on what we think should happen in our AI governance primer. In short, we think that while there are many reasons to be excited about AI, we also urgently need to build more infrastructure for guardrails, security, and democratic control before we can really trust the technology to deliver for the common good.)
Best case: The Trump administration takes accident and misuse risks from AI seriously, preserving relevant portions of the Biden Executive Order that address those risks or replacing them with alternatives that are at least as good. The US AI Safety Institute or a functional equivalent is preserved and funded by Congress, and leading labs have their frontier models audited by the government prior to release. The administration takes care to avoid unnecessary escalation with China and does not turbocharge new investments in military or domestic state surveillance applications of AI. The government invests in and/or incentivizes socially positive AI applications such as for medical research, and creates public infrastructure to support that research. The administration participates in global AI governance conversations and does not block international collaboration initiatives led by other countries. The administration and Congressional Republicans do not pick high-profile fights with algorithmic bias and misinformation researchers/advocates, either working with them behind the scenes on areas of overlapping interest or letting them pursue their objectives outside of the federal government sphere.
Worst case: The Trump administration goes full accelerationist, directing massive investment to support military and national security applications of artificial general intelligence (AGI), making a new arms race with China a self-fulfilling prophecy. The hysteria associated with the racing narrative greatly increases xenophobia and the risk of armed conflict even as it incentivizes corner-cutting, corruption, and criminal activity across every component of the AGI supply chain. Meanwhile, the administration and Congressional Republicans adopt a laissez-faire approach to commercial development of AI that provides no protection against misuse at scale and loss of control to AI systems. A lack of regulatory constraints empowers leading AI developers and suppliers to engage in widespread rent-seeking behavior; social inequality increases as value-destroying applications proliferate. As a “Manhattan Project for AGI” proceeds and AI technology continues to become deeply integrated into American life and critical infrastructure, a government-corporate oligarchy develops among the executive branch and executives of leading labs, investors, and defense contractors, threatening institutional independence and the rule of law. AI development becomes politically polarized, with anyone arguing for restraint via domestic regulations or international diplomacy branded as anti-American.
Our current forecast: 5/10 (↓1). As expected, the administration canceled the Biden executive order on day one, but shortly afterwards issued a directive for National Security Adviser Mike Waltz, OSTP head Michael Kratsios, and AI czar David Sacks to come up with a new plan within 180 days. The fact that a detailed directive wasn’t already in place suggests that the administration hasn’t yet settled on a strong direction for AI governance and there is room for a range of voices to have input. Similarly, while the US AI Safety Institute’s director Elizabeth Kelly stepped down in February, USAISI itself doesn’t seem to have been targeted for elimination and indeed is getting some backup from allies of the administration. No concrete plans for an AGI Manhattan Project have taken hold publicly, but we are dropping one point because of JD Vance’s accelerationist rhetoric at the Paris AI Action Summit and the lack of interest it signaled in multilateral cooperation on AI governance. Last updated: March 2025
National security and foreign policy
(Peace and security is a core theme for EIP. Generally, our philosophy is simple: conflict is bad, peace good — with some allowances for the perverse incentives of failing to hold bad actors accountable for territory grabs and such. For the most part, we are focused on preventing and mitigating very large-scale conflicts, such as nuclear war or war between great powers.)
Best case: The Trump administration’s strategic ambiguity keeps adversaries guessing and incentivizes solicitous behavior, as in the case of North Korea during Trump’s first term. US policy towards Ukraine and Israel succeeds in ending wars on not-too-unfavorable terms to victimized civilian populations, stopping fatalities in the short term and making renewed armed conflict less likely in the medium term. A nuclear buildup is limited and Trump makes clear his distaste for using nuclear weapons, clearing the way for renewed Track 1 arms control negotiations between nuclear powers. Trump does not meaningfully divest from NATO or the multilateral peacekeeping system, preventing an erosion of solidarity among Western democracies.
Worst case: The Trump administration’s aggressive stance towards China greatly increases the risk of armed conflict in the region, or even directly leads to such conflict. Trump withdraws support for Ukraine and weakens the US’s commitment to NATO, allowing Russia to take effective control of the country and emboldening Putin to threaten other neighbors. The US’s combined policy towards Russia and China is incoherent and contradictory, simultaneously encouraging the spread of authoritarianism around the world while continuing to escalate tensions in the South China Sea. Tariff policy creates tensions with allies and adversaries alike and isolates the US on the global stage. The US disengages from the multilateral system, triggering a funding crisis at the UN and reducing international capacity to prevent and mitigate coups, internal conflicts, and regional wars around the world.
Our current forecast: 5/10 (↓1). Trump’s approach to foreign policy so far has been at times a shocking departure from past US leaders, but it’s still a little early to judge how it will ultimately play out for global peace and security. The clearest negative signal is that Trump is back to embracing Russian talking points and this time around seems intent on following through with a realignment of US-Russia relations that will likely harm war victims and erode the taboos against violating territorial sovereignty, a trend exacerbated by the administration’s clear skepticism of the United Nations and bids to withdraw funding from various organs of the multilateral system. Perversely, though, Trump and Putin’s mutual affinity might go a long way toward reducing the risk of nuclear conflict, at least in the short term. Last updated: March 2025
Global health
(EIP’s biggest priorities within global health include biosecurity / pandemic preparedness, medical research and accessibility for life-saving vaccines and treatments, reducing air pollution, and interventions for neglected tropical diseases. We consider the US public health system to be included in our definition of global health and so changes to it would also fall in this category.)
Best case: The Trump administration takes pandemic risk at least somewhat seriously and invests in reasonable safeguards, such as PPE stockpiles and restricting gain-of-function research. There is no degradation in current international agreements against bioweapons and the administration participates in relatively good faith in international negotiations such as the WHO pandemic treaty. The United States continues to be a major financial supporter of the WHO, Gavi, and The Global Fund to Fight AIDS, Tuberculosis, and Malaria, and Congress largely preserves the most net beneficial global health aid expenditures. New rules designed to remove barriers to scientific innovation spur faster advances in medical research and rollout of beneficial medications and other technologies. Domestically, government officials continue to promote use of vaccines and other treatments based on credible science, and initiatives such as the Department of Government Efficiency find ways to reduce government health spending without adversely affecting patient outcomes.
Worst case: The Trump administration withdraws from the WHO, severely destabilizing it and threatening its future viability due to lack of funding. Tensions with geopolitical rivals such as China spill over into the global health sphere, rehashing arguments over the origins of COVID and harming international cooperation to combat future pandemics. The United States torpedoes any possibility of a meaningful international pandemic preparedness agreement in the near term. Congress imposes major cuts to US global health aid and allows PEPFAR to expire. Statements and policy changes by government officials greatly reduce vaccine uptake and diseases such as measles begin to make a comeback in the United States, just as a steady talent drain leaves the FDA and CDC ill-equipped to respond to future disease outbreaks. Health policy begins to become untethered from science as a cottage industry of medical conflict entrepreneurs and conspiracy theorists coalesces to influence administration and Congressional actions via conservative media.
Our current forecast: : 2/10 (↓3). What’s happened to date on the global aid front is actually worse than the worst-case scenario we described above: right out of the gate, the government ordered a halt to all foreign assistance, including one of the most cost-effective global health programs on the planet, and subsequently canceled thousands of contracts that underpinned critical infrastructure for the entire humanitarian assistance field worldwide. Nearly all USAID employees have been placed on administrative leave and told to collect their belongings as the administration seeks to fold the agency into the Department of State. Oh, and the administration did order a withdrawal from the WHO. Domestically, Robert F. Kennedy Jr.’s Department of Health and Human Services isn’t inspiring confidence with its tepid response to the current vaccine-hesitancy-driven measles outbreak. Last updated: March 2025
Democracy and institutional accountability
(According to the research on wellbeing, living in a democracy has positive effects on life satisfaction after controlling for other factors, and a similar correlation exists for perceived freedom, which we construe as connected to civil liberties such as freedom of expression and political choice. Trump’s well-documented authoritarian tendencies have been a longstanding source of concern for those who have opposed him politically.)
Best case: Motivated by a desire to win the respect of his political opponents and be celebrated by future historians, Trump takes a magnanimous view towards the immense power he holds. Throughout the term, his administration follows court orders, respects longstanding norms around separation of powers (aside from those it has explicitly and repeatedly promised to overturn), and is more focused on the present and future than relitigating the past. Politically motivated prosecutions are limited and there is no appreciable increase in state surveillance of American citizens or use of government power to coerce fealty to Trump and his priorities. Trump nominates well-qualified conservative judges, at least some of whom have a track record of judicial independence. Moderate/institutionally-minded Republicans in the House and Senate exercise independence and erect some guardrails on abuses of power by the executive. Actions to deport undocumented immigrants are limited in scale and with a minimum of drama, and the administration does not engage in large-scale harassment of other vulnerable citizens.
Worst case: Driven by Trump’s own obsessions, the administration becomes consumed with retribution against its perceived enemies. Trump treats the DOJ as an extension of his personal legal team and uses it to prosecute and threaten political opponents, delegitimize the 2020 election results (and future election results), and insulate allies against legal accountability for their actions. Domestic intelligence and law enforcement agencies are regularly deployed for the president’s personal and political gain, and the White House uses its budget authority to try to blackmail agencies and state governments into doing its bidding, with no political will in Congress or the courts to stop any of this. The administration uses a revitalized Schedule F executive order to fire massive numbers of civil servants and hollow out the system of merit-based hiring across the executive branch. Trump sends the Senate a steady stream of judicial appointments selected only for their loyalty, each of whom are confirmed no matter how unqualified or unfit. A drive to eliminate “wokeness” from the military serves as cover for a leadership purge, undermining the professionalism of and bipartisan trust in the military. Under intense pressure, Senate Republicans eliminate the filibuster and pass sweeping legislation reshaping voting rules and redistricting processes that give the GOP a virtually insurmountable advantage in future Congressional elections. The administration engages in clear and ongoing violations of basic human rights in the course of setting up high-profile deportation camps for undocumented immigrants, with no repercussions or accountability either domestically or internationally, and pursues policies seemingly aimed at bullying and scapegoating vulnerable members of society.
Our current forecast: 2/10 (↓2). The first two months of the administration have made clear Trump’s eagerness to deploy the power of the state not only to settle old scores but also to instigate new ones. The retribution campaign started on day one with the mass clemency for every single January 6 protester, including those convicted of violent crimes, and rapidly extended to the firing of prosecutors and investigators involved with those cases. More recently, the administration has undertaken retaliatory actions against private sector associates of those involved with prosecutions of Trump during his time as a private citizen. Just about every nonpartisan agency or appointment has seen its officials fired or operations reorganized, from the National Archivist position to the White House Office of Government Ethics to top lawyers in the military. Trump has talked openly about withholding federal funding to Democratic states that oppose his objectives, and his jokes about running for an unconstitutional third term are getting more frequent. While the filibuster is holding strong for now, it might not matter much in the end: between the administration’s embrace of impoundments and seeming willingness to ignore court orders, Congress’s practical ability to hold the executive branch in check is growing weaker by the day. Last updated: March 2025
State capacity and institutional functioning
(The government’s ability to realize its goals while respecting Constitutional norms is critical for all life-improving outcomes it may be able to deliver. We are supportive of interventions like addressing operational bottlenecks within Congress and other key institutions, electoral reforms to shift incentives for lawmakers, and enabling the government to attract and retain the best talent.)
Best case: The Department of Government Efficiency, led by tech entrepreneurs Elon Musk and Vivek Ramaswamy, identifies a number of legitimate ideas to streamline government operations that can be enacted legally while sparing the most important components of the executive branch that are making good use of existing capacity. The continued narrow majority in the House emboldens moderate Republicans to flex their muscle, incentivizing leadership to focus on governing over petty spats and spurring more policy entrepreneurship across fluid cross-partisan coalitions. The primary reform movement retains momentum after successfully defending Alaska’s final-four voting system, securing at least one high-stakes statewide victory in the 2026 cycle. A bipartisan majority in Congress reauthorizes the Defense Production Act and enacts other incremental reforms to preserve or improve the capacity of the state to deliver beneficial results for constituents.
Worst case: DOGE recommends overly aggressive and poorly targeted cuts to government functions that either get bogged down in the legal system or are imposed extralegally, provoking a Constitutional crisis. Partisan rancor consumes Congress daily and pushes out any possibility of meaningful reforms, changes in culture, or defense against abuses of executive power. DPA reauthorization fails or gets loaded up with provisions intended to authorize questionable uses of its powers by the president. Primary and other electoral reform efforts stall out as a result of the failure of high-profile ballot initiatives in 2024. The use of Schedule F hollows out the civil service and makes the executive branch even less effective than it was before.
Our current forecast: 4/10 (↓1). Perhaps nothing has better demonstrated the high-variance nature of the Trump administration than the rapid growth and evolution of DOGE. With Ramaswamy out of the picture, Musk’s handpicked team has burrowed deep into the federal government, taking over both the human resources system for all domestic operations and the payments system for the entire government, including the military. It has used this authority to force drastic changes in the way that agencies operate, most notably at USAID which was effectively shuttered over the course of a single weekend, but less dramatically across almost every area of government. On the one hand, in some ways there has been a definite increase in state capacity as a result of these efforts: Musk’s team, for example, built the capability to mass-email federal employees messages like the “Fork in the Road” offer, and there are rumors that the DOGE team is laying the groundwork for mass automation of the federal government. But our best-case scenario had stipulated that these reforms would be “enacted legally while sparing the most important components of the executive branch that are making good use of existing capacity,” which emphatically has not described the most important events of the past month. Last updated: March 2025
Where Time and Attention Seem Most Valuable Right Now
Considering the above scenarios in combination with what we’re hearing on the ground from colleagues in and around government, we are drawing the following preliminary conclusions about interventions to consider in the coming years:
With high confidence, we believe that advocating for and trying to restore the most valuable components of the decimated US foreign assistance regime merits attention and high priority. Foremost among these is the President’s Emergency Plan for AIDS Relief (PEPFAR), which until recently was almost certainly the most cost-effective activity undertaken by the US government in terms of lifesaving benefit per dollar spent, comparable to top GiveWell charities. But several other USAID and State Department programs were critical to those highly effective charities being able to operate, including the President’s Malaria Initiative and contractor systems that were critical to the humanitarian assistance supply chain, and USAID had also supported several standout charities like GiveDirectly, well, directly. Several bridge funds have popped up to plug immediate needs in this ecosystem, with tens of millions of dollars in identified gaps, and there are also emerging efforts to coordinate advocacy for a longer-term solution to this crisis that is capable of attracting bipartisan buy-in.
With moderately high confidence, we believe that guiding AI policy is a promising opportunity in the current administration. The Trump administration’s approach to AI policy appears to be quite fluid and open to input right now, and the next several years may well be decisive for the trajectory of US AI regulation and by extension global AI governance. We believe the most important and likely failure mode of Trump’s approach to AI policy is to be too accelerationist and willing to cut corners on safety for the sake of competing with China, but this outcome is far from predetermined both in the short and medium term. While some sort of race for dominance with China looks inevitable at this point, the question of whether the race stays “manageable” is very much still up for grabs.
With moderately high confidence, we believe that improving epistemics around US-China relations merits high priority. By default, the US is likely to adopt an adversarial posture vis-a-vis China due to the prevalence of national security hawks in the Republican Party and the incentives of politically well-connected defense technology companies and their investors. Rather than an inherently hawkish or dovish position, EIP believes that US-China relations should be based on facts and realistic readings of China’s capabilities and aims; in other words, we think the most important negative outcomes to avoid are (in)actions with irrevocable consequences that are based on fundamental miscalculations of the bilateral relationship. To avoid such outcomes, continued investment in Track 1.5/2 dialogues and building risk-limiting infrastructure such as nuclear hotlines are important. Currently, the civil society organizations that have been most active arranging such dialogues are not well connected with the right-leaning national security community, so better integrating these networks strikes us as a robustly good objective.
With moderate confidence, we believe that protecting core institutional accountability values such as freedom of speech and judicial independence is an important objective in this administration. Because the administration is seeking judicial sanction for a dramatic expansion of power for the executive branch, funding lawsuits to challenge the most troubling uses of that power – and to crowd in legal talent to carry forward those lawsuits – may be an especially pivotal use of funding in the next few months. We also think that the DC political bureaus of key newsrooms are performing an important public service by tracking administration actions and the implications of them. For smaller donors in particular, subscriptions to the New York Times, Washington Post, and other outlets currently offering high-quality investigative reporting can be an important way to insulate those newsrooms from corporate interference and pressure from lawsuits.
With weak-to-moderate confidence, we believe that guiding the US DOGE Service is a promising opportunity. Despite its gimmicky name and chaotic implementation, DOGE is (albeit roughly) aimed at a real problem: the fossilization of large portions of the public sector, including the defense-industrial complex, under the weight of regulatory bloat. DOGE’s whole-of-government scope and blank-slate mandate create a wide space of possibilities that could kneecap the government’s most valuable work or yield some of the biggest achievements of the Trump administration over the next four years. As DOGE staff and new agency heads get past the destruction stage and become more familiar with how government works, more coherent organizational visions and plans will start to emerge; finding ways into those design conversations, while far from straightforward, could be highly valuable.
As noted above, we will add notes on further focus areas as the year proceeds.
Learn More and Get Involved
Helping donors and institutional reformers navigate the new reality in DC is a major focus for EIP in 2025. Every day, we are monitoring the news and speaking to experts inside and outside of government to stay on top of the administration’s evolving priorities. We are offering 1:1 calls and tailored strategic advice for funders who wish to take action on these issues. If you are interested in partnering, please fill out this form.
This is a living document that we will update throughout 2025, if not longer. We expect to update our forecasts at least quarterly, and other material as needed/appropriate. The most recent revision date will be noted at the top. Please contact info@effectiveinstitutionsproject.org or comment below with any feedback, tips, or suggestions.